Studies with potential reporter group reagents for enzymes: 3,4-Dihydro- 3-(2-hydroxyethyl)-6-nitro-2H-1,3-benzothiazin-2-thione and 6-nitrochromone hydrolysis and interaction with chymotrypsin

Article abstract of DOI:10.1006/bioo.2000.1164

3,4-Dihydro-3-(2-hydroxyethyl-6-nitro-2H-1,3-benzothiazin-2-thione (II- HE), a cyclic dithiocarbamate, has the potential to react with an esterase to furnish it with a p-nitroben-zenethiolate 'reporter group.' However, unlike a closely similar cyclic carbamate, II-HE is totally without reaction on chymotrypsin. Possible reasons for this major effect of substituting sulfur for oxygen are discussed. The results support the idea that chymotrypsin's 'oxyanion hole' cannot properly accommodate a thioanion. II-HE undergoes an interesting intramolecular cleavage reaction under alkaline conditions. The mechanism of this process has been determined using evidence from NMR and mass spectrometry. 6-Nitrochromone (6-NC) likewise has the potential to modify an enzyme Covalently and thereby act as a reporter group reagent. With chymotrypsin, 6-NC reacts as predicted, except that the attached label is stable only at high pH; the labeling reaction slowly reverses at low pH. From the lack of effect on enzyme activity, it is clear that modification of chymotrypsin by 6-NC does not occur at the active site. (C) 2000 Academic Press.

Full text of DOI:10.1006/bioo.2000.1164

Bioorganic Chemistry 27, 73–88 (2000)
doi:10.1006/bioo.2000.1164, available online at http://www.idealibrary.com on
Studies with Potential Reporter Group Reagents for Enzymes:
3,4-Dihydro-3-(2-Hydroxyethyl)-6-Nitro-2H-1,3-Benzothiazin-2-
Thione and 6-Nitrochromone—Hydrolysis
and Interaction with Chymotrypsin
Trevor M. Kitson
Institute of Fundamental Sciences, Massey University, Palmerston North, New Zealand
Received August 5, 1999
3,4-Dihydro-3-(2-hydroxyethyl)-6-nitro-2H-1,3-benzothiazin-2-thione (II-HE), a cyclic
dithiocarbamate, has the potential to react with an esterase to furnish it with a p-nitroben-
zenethiolate “reporter group.” However, unlike a closely similar cyclic carbamate, II-HE is
totally without reaction on chymotrypsin. Possible reasons for this major effect of substi-
tuting sulfur for oxygen are discussed. The results support the idea that chymotrypsin’s
“oxyanion hole” cannot properly accommodate a thioanion. II-HE undergoes an interesting
intramolecular cleavage reaction under alkaline conditions. The mechanism of this process
has been determined using evidence from NMR and mass spectrometry. 6-Nitrochromone
(6-NC) likewise has the potential to modify an enzyme covalently and thereby act as a
reporter group reagent. With chymotrypsin, 6-NC reacts as predicted, except that the
attached label is stable only at high pH; the labeling reaction slowly reverses at low pH.
From the lack of effect on enzyme activity, it is clear that modification of chymotrypsin by
6-NC does not occur at the active site.
© 2000 Academic Press
INTRODUCTION
In previous work, the cyclic carbamate 3,4-dihydro-3-methyl-6-nitro-2H-1,
3-benzoxazin-2-one (I)1 was used to react with enzymes having an esterase function
(including chymotrypsin and aldehyde dehydrogenase) in order to provide them with a
coloured p-nitrophenoxy “reporter group” (1–3). Scheme 1 shows the chemistry of this
process. The first part of the present work concerns analogous compounds with a dithio-
carbamate instead of a carbamate group. It was thought that such a compound might
either react with an esterase in just the same way as I (providing an alternative and more
easily synthesised reporter group reagent) or it might show an interestingly different
chemistry to that of I. Accordingly, the synthesis is reported here of 3,4-dihydro-3-
1 Abbreviations used: I, 3,4-dihydro-3-methyl-6-nitro-2H-1,3-benzoxazin-2-one; II-Me, 3,4-dihydro-3-
methyl-6-nitro-2H-1,3-benzothiazin-2-thione; II-HE, 3,4-dihydro-3-(2-hydroxyethyl)-6-nitro-2H-1,3-
benzothiazin-2-thione; 6-NC, 6-nitrochromone. The abbreviations II-Me and II-HE draw attention to the
relevant sidechains, methyl and hydroxyethyl, respectively.
73
0045-2068/00 $35.00
Copyright © 2000 by Academic Press
All rights of reproduction in any form reserved.

74
TREVOR M. KITSON
SCHEME 1. Reaction of I with an enzymic nucleophile to provide a p-nitrophenoxy reporter group.
methyl-6-nitro-2H-1,3-benzothiazin-2-thione (II-Me) and 3,4-dihydro-3-(2-hydro-
xyethyl)-6-nitro-2H-1,3-benzothiazin-2-thione (II-HE); the route is shown in Scheme 2.
The advantage of this synthesis over that of I (1) is that it is shorter and does not
involve the dangerously toxic phosgene. The reactivity of these compounds toward chy-
motrypsin and to alkaline hydrolysis is described below.
The second part of the present work concerns an investigation of 6-nitrochromone
(6-NC, 6-nitro-4H-1-benzopyran-4-one) as another potential reporter group reagent.
This compound is of very similar size and shape to I and in principle could react with
an enzymic nucleophile as shown in Scheme 3, again incorporating a covalently linked
p-nitrophenoxy group. Chromone itself is known to react in similar fashion with diethy-
lamine, giving 3-diethylamino-1-(2-hydroxyphenyl)propenone in good yield (4). 6-NC
is very simply prepared by the nitration of chromone; its reaction with chymotrypsin is
described below.
MATERIALS AND METHODS
Preparation of II-HE (See Scheme 2)
2-Chloro-5-nitrobenzaldehyde (4.0 g) was dissolved in ethanol (40 ml) and an equi-
molar amount of 2-aminoethanol was added; the mixture was left overnight at room
temperature. The ethanol was evaporated and the residue was dissolved in chloroform;
this solution was then washed with water (to remove any unreacted aminoethanol), dried

REPORTER GROUP REAGENTS FOR ENZYMES
75
SCHEME 2. Synthetic route leading to cyclic dithiocarbamates. II-Me, R = CH3; II-HE, R =
CH2CH2OH.
over MgSO4, and evaporated to give a viscous yellow oil. The oil was dissolved in
ethanol (50 ml) and cooled in ice. Sodium borohydride (0.8 g) was added slowly in por-
tions and the resulting orange solution was left overnight at room temperature. The
ethanol was evaporated and the residue partitioned between chloroform and water.
The chloroform layer was dried over MgSO4 and evaporated to dryness. The resulting
oil crystallized to a pale orange solid. The yield of N-(2-hydroxyethyl)-2-chloro-5-
nitrobenzylamine was 4.07 g (82% over the two steps). δ (CDCl3): 8.36 (d, 1H), 8.10
(d of d, 1H), 7.53 (d, 1H), 3.99 (s, 2H), 3.74 (t, 2H), 2.85 (t, 2H), 2.02 (s, 2H, NH and
OH). This material was dissolved in ethanol (40 ml) and equimolar amounts of tri-
ethylamine and carbon disulfide were added; the mixture was left overnight at room
temperature. After removal of solvent, the residue was extracted into chloroform and

76
TREVOR M. KITSON
SCHEME 3. Nucleophilic attack on 6-NC. X− = HO− or represents an enzymic nucleophilic group.
water; the chloroform layer was dried over MgSO4 and evaporated to dryness. The
resulting red solid was dissolved in a small amount of warm chloroform and applied
to a column of silica gel, eluting with chloroform. The major fraction that eluted from
the column was evaporated to give a yellow solid. The yield of 3,4-dihydro-3-(2-
hydroxyethyl)-6-nitro-2H-1,3-benzothiazin-2-thione was 1.4 g (29%); the NMR spec-
trum showed the compound to be pure. δ (CDCl3): 8.17 (d, 1H), 8.22 (d of d, 1H), 7.41
(d, 1H), 4.83 (s, 2H), 4.32 (t, 2H), 4.05 (t, 2H), 1.9 (s, 1H).
Preparation of II-Me
This compound was prepared by essentially the same method as described above
for II-HE, except methylamine (as a 40% aqueous solution) was used instead of
2-aminoethanol. δ (CDCl3): 8.17 (d, 1H), 8.21 (d of d, 1H), 7.38 (d, 1H), 4.74 (s, 2H),
3.66 (s, 3H). m/z = 240.002239 (calcd for C9H8N2O2S2: 240.002721).
Preparation of 6-NC
Chromone was nitrated by a published method (5); the m.p. and NMR spectrum of
the product were the same as in the literature.
Spectrophotometry
All UV/visible spectra were recorded using a Varian Cary 1 instrument. The activity
of chymotrypsin was monitored using p-nitrophenyl acetate (0.1 or 0.25 mM) as sub-
strate in Tris (50 mM, pH 8.5) or phosphate buffer (50 mM, pH 8.0) at 25°C. Stopped-
flow experiments were carried out with a Hi-Tech Scientific instrument using
p-nitrophenyl acetate (0.25 mM) as substrate in 50 mM phosphate buffer, pH 8.0, at
25°C.
Modification of Chymotrypsin
II-HE or I in acetonitrile (0.2 ml) was added to 0.2 M phosphate buffer (pH 8.0,
2.8 ml) containing chymotrypsin (α-chymotrypsin, type II from bovine pancreas,

REPORTER GROUP REAGENTS FOR ENZYMES
77
Aldrich); the resulting concentrations of enzyme and modifier were 0.22 and 0.5 mM,
respectively. The mixture was protected from light and left at 29°C for 45 h. Samples
were taken at intervals for enzyme assay. The mixture was then passed down a gel fil-
tration column (Biogel P-6, 25 × 0.8 cm), eluting with 10 mM phosphate buffer, pH 7.4,
and after making the solution alkaline the protein fraction was scanned spectropho-
tometrically. 6-NC in acetonitrile (0.2 ml) was added to 50 mM phosphate buffer
(pH 8.0, 4.8 ml) containing chymotrypsin; the resulting concentrations of enzyme and
modifier were 0.25 and 0.5 mM, respectively. The mixture was left at room tempera-
ture overnight and then subjected to gel filtration using 10 mM phosphate (pH 6.0) as
eluant. Samples (0.5 ml) of the protein fraction were added to 2 ml of various more
concentrated buffers; the resulting pH was measured and the solutions were scanned
spectrophotometrically.
Hydrolysis Product of II-HE
II-HE (5.76 mg) was dissolved in acetonitrile (2 ml) and added to 0.1 M NaOH
(200 ml) at room temperature. After 5 min, the yellow solution was acidified by adding
1.9 ml of concentrated hydrochloric acid (specific gravity 1.18) and then extracted with
chloroform (25 ml). The chloroform solution was dried over MgSO4 and evaporated to
dryness under high vacuum. The residue was dissolved in CDCl3 for NMR spectrome-
try and also subjected to mass spectrometry. m/z = 270.011548 (calcd for C10H10N2O3S2:
270.013286), 272.002052 (calcd for C10H9N2O3S35Cl: 272.002242), 274.000076 (calcd
for C10H9N2O3S37Cl: 273.999292).
Treatment of II-HE with Perchloric Acid
II-HE (20 mg) was dissolved in chloroform and perchloric acid (70%, 8.0 µl) was
added. The mixture was left overnight at room temperature and then evaporated to
dryness. The residue was dissolved in d6-DMSO for NMR spectrometry.
RESULTS AND DISCUSSION
Synthesis of Cyclic Dithiocarbamates
The compounds II-Me and II-HE were successfully prepared by the route shown
in Scheme 2, albeit in low yield for the cyclisation step. No doubt an extensive explo-
ration of the reaction conditions would allow the yield to be optimized, but this was
not considered necessary for the present work as only very small amounts of the
reagents are required for enzyme experiments. The route involves a nucleophilic aro-
matic substitution reaction that occurs under unusually mild conditions; the dithiocar-
bamate ion is ideally positioned for intramolecular attack, and the nitro group provides
an electron sink to stabilise the intermediate. Using the nitro group in this way could
be described as “chemical parsimony,” since the nitro group already has the essential
functions (in the reaction for which these reagents are designed) of enhancing the
leaving group ability of the benzenethiolate ion and of rendering it colored.
II-HE and Chymotrypsin
Unfortunately II-Me (the simple dithio analogue of I) was found to be too insolu-
ble in water to use as a potential enzyme-modifying reagent. This was the reason why

78
TREVOR M. KITSON
the hydroxyethyl derivative (II-HE) was also synthesized, and as expected its solu-
bility is much higher.
When chymotrypsin (0.22 mM) was incubated at 29°C and pH 8.0 with I (0.50 mM),
there was substantial loss of enzyme activity (60% inactivation in 25 h). However,
under the same conditions, the activity of the enzyme in the presence of II-HE was
128% of the control incubation without reagent. Upon isolation of the I-modified
enzyme by gel filtration (and after making the solution alkaline with a small volume
of NaOH solution), the UV/visible spectrum shown in Fig. 1a was obtained. It can be
calculated from the magnitude of the absorbance that approximately 61% of the enzyme
is covalently labeled by I. (These results are similar to those of previous work; 1) When
treated in exactly the same way, the solution of enzyme that had been incubated with
II-HE gave the spectrum shown in Fig. 1b; clearly there is no trace at all of any reac-
tion between II-HE and chymotrypsin. (Presumably the higher level of enzyme activ-
ity after incubation with II-HE is due to the compound binding noncovalently in the
active site and protecting the enzyme somewhat from the appreciable spontaneous
denaturation or autolysis that occurs over a long time period.)
It seems conceivable but unlikely that the different behavior of chymotrypsin
toward II-HE and I is due to a steric effect of replacing I’s methyl group by the only
slightly larger and conformationally flexible hydroxyethyl sidechain. This conclusion
is strongly supported by the fact that chymotrypsin reacts avidly at the ring carbonyl
group of the compound shown (along with II-HE for structural comparison) in Scheme
4 (6). Rather, the different outcomes may reflect a major difference in chemical reac-
tivity of a dithiocarbamate compared to a carbamate (and evidence on this point is pre-
sented in the next section). Dithiocarbamates are reported to be fairly unreactive (7);
for example, treatment of Me2NCS-SCH2COOMe with ammonia at 100°C gives
Me2NCS-SCH2-CONH2, in which the dithiocarbamate group is unchanged. However,
the simple dithiocarbamate Et2NCS-SMe is by no means totally inert under mild con-
ditions; it is oxidized in vivo first to Et2NCO-SMe and then to Et2NCO-SOMe (8).
Moreover, tetramethylthiuram monosulfide (Me2NCS-S-CSNMe2) reacts with a vari-
ety of nucleophiles (RO−, RS−, amines; 9) in just the manner in which it was envisaged
FIG. 1. (a) The UV/visible spectrum of the reporter group (at alkaline pH) supplied to chymotrypsin
by I. (b) The equivalent result obtained when chymotrypsin was treated with II-HE under exactly the same
conditions as used with I in (a).

REPORTER GROUP REAGENTS FOR ENZYMES
79
SCHEME 4. Comparison of the structure of a known inactivator of chymotrypsin (6) with that of
II-HE.
that II-HE might react with an enzyme (in this case the leaving group being the dime-
thyldithiocarbamate ion rather than a p-nitrobenzenethiolate ion).
An observation that is relevant to the present results is that chymotrypsin does not
measurably catalyze the hydrolysis of thionoesters (substrate analogues with a thio-
carbonyl group), although thioloesters (in which the singly bonded oxygen of an ester
group is replaced by sulfur) are good substrates (10,11). It was suggested from this
that the enzyme’s “oxyanion hole” is very specific for oxygen and cannot stabilize the
tetrahedral transition state when sulfur is involved. Interestingly, cysteine proteases
such as papain do act on thionoesters (12). Thus II-HE may prove to be a useful
reporter group reagent for some enzymes (other than serine proteases); in particular,
it seems likely to react with aldehyde dehydrogenase (which has been extensively
studied using I: 2,3) since this enzyme is known to be inactivated by a noncyclic
dithiocarbamate analogue of II-HE, p-nitrophenyl dimethyldithiocarbamate (13).
Alkaline Hydrolysis of I, II-Me, and II-HE
The difference in reactivity of a carbamate and a dithiocarbamate is clearly demon-
strated by adding solutions of I and II-Me (in acetonitrile) to 0.1 M NaOH. In the case
of the former, hydrolysis occurs rapidly giving a yellow solution with λmax = 411 nm
and ε = 18,200, as expected. However, with II-Me the initial absorbance at 320–330 nm
slowly declines (losing about half its intensity in 20–30 min), but with no development
of any absorbance in the visible region whatsoever. This is evidently due to the very
poorly soluble II-Me slowly being deposited from the originally 50 µM solution, but
without undergoing any hydrolysis. (This phenomenon happens in any aqueous solu-
tion, not just 0.1 M NaOH.) Oxygen and sulfur differ in terms of size, polarizability,
stability of their multiple bonds, etc., but perhaps the simplest way of rationalizing the
observed difference in reactivity of the carbamate and the dithiocarbamate is to note
that in the former the carbon atom is connected to two highly electronegative oxygen
atoms and is therefore more electrophilic than in the latter. In interesting contrast to
II-Me, the hydroxyethyl analogue II-HE quickly gives a yellow solution when added
to 0.1 NaOH; the λmax is coincidentally the same as that of the hydrolysis product of I

80
TREVOR M. KITSON
(411 nm), but the absorbance peak is somewhat broader and has a lower molar absorp-
tivity (ε = 12,400).
It is quite clear from the observations just described that the neighboring hydroxyl
group of II-HE provides anchimeric assistance in some way to the alkaline cleavage of
the dithiocarbamate group. The most obvious mechanism by which this might occur is
shown in Scheme 5. Here it is envisaged that although the dithiocarbamate group is not
productively attacked by hydroxide ion intermolecularly (as shown by the inertness of
II-Me), it is perfectly capable of being cleaved if the attacking nucleophile is held in
close proximity to it. However, there is an alternative mechanism for the hydrolysis of
II-HE, as shown in Scheme 6, in which the dithiocarbamate acts as the nucleophilic
rather than the electrophilic component. This route involves the initial displacement of
hydroxide ion, a poor leaving group, but again the close proximity of the neighboring
(dithiocarbamate) group may render the process favourable. Certainly a very similar
reaction occurs with a better leaving group, as in Me2NCS-SCH2CH2Cl, and there is
some evidence that it can happen in Me2NCS-SCH2CH2OH as well (13). In Scheme 6,
the resulting dithioiminium ion is then attacked by hydroxide ion as shown leading to
the release of the nitrobenzenethiolate moiety. The key steps in Schemes 5 and 6 are
defined as 5-exo-trig and 5-exo-tet, respectively, and thus are both allowed by Baldwin’s
rules. The products shown in the schemes both possess a cyclic thiocarbamate group, but
SCHEME 5. A proposed mechanism for the alkaline cleavage of II-HE.

REPORTER GROUP REAGENTS FOR ENZYMES
81
SCHEME 6. An alternative proposed mechanism for the alkaline cleavage of II-HE.
differ as to whether the oxygen or the sulfur is in the ring or outside it. It was decided to
isolate the hydrolysis product of II-HE and characterise it by NMR spectroscopy in
order to elucidate which of Schemes 5 and 6 actually occurs.
II-HE was subjected to alkaline hydrolysis and, after acidification of the reaction mix-
ture with hydrochloric acid, the product was extracted into chloroform and its NMR
spectrum was determined; see Fig. 2. The large peak at 7.27 ppm is due to chloroform.
The small peaks at 4.83, 4.32, and 4.05 ppm are due to residual unhydrolyzed II-HE.
The spectrum clearly shows the presence of two main compounds of similar structure
(and integration suggests they are present in a ratio of 57Ϻ43). For example, regardless
of whether Scheme 5 or 6 operates, we would expect the product to show one singlet
(the benzyl CH2), but there are two (near 5 ppm). We would expect one doublet from
the aromatic proton ortho to the sulfur, but again there are two doublets (at 7.83 and
7.48 ppm). (The other aromatic signals are poorly resolved at about 8 to 8.2 ppm.) We
would expect two triplets from the adjacent CH2 groups, but there are two close togeth-
er at about 3.7 ppm and two more appear to be overlaid at about 4.6 ppm. The latter res-
onances must be due to CH2O groups in both components of the mixture (and the peaks
near 3.7 due to CH2N groups); critically there is no sign of any resonance in the region
expected for a CH2S group. For instance, the CH2 group in Me2N-CO-SCH2CH3 has a

82
TREVOR M. KITSON

REPORTER GROUP REAGENTS FOR ENZYMES
83
signal at 2.87 ppm (15), whereas that in Me2N-CS-OCH2CH3 appears at 4.40 ppm (16).
The cyclic analogue of the latter compound, 3-methyl-1,3-oxazolidin-2-thione (which is
a good model for the product of Scheme 5), shows triplets at 3.90 and 4.52 correspond-
ing to the methylene groups next to nitrogen and oxygen, respectively (17). Thus we can
conclude with confidence that Scheme 6 does not apply, and that the major component
of the reaction mixture is that shown in Scheme 5 (after protonation). We know it is the
major component since the larger of the two doublets in the region from 7 to 8 ppm has
a chemical shift that is more compatible with a proton ortho to a neutral sulfur sub-
stituent; II-HE itself for instance has a doublet at virtually the same position, 7.41 ppm.
(The other doublet is discussed below.)
To gain further information about the identity of the minor product, the mixture was
analyzed by mass spectrometry. A molecular ion at 270 was observed, confirming the
presence of the product shown in Scheme 5 (after protonation). Large peaks were also
seen at 272 and 274 in a ratio of approximately 3:1, strongly suggestive of the presence
of chlorine. High resolution mass spectrometry confirmed the formula C10H9N2O3SCl.
Thus the minor product is related to the major by loss of SH and gain of Cl; two plausi-
ble mechanisms for how this might happen when the initial hydrolysis solution is acid-
ified with HCl are shown in Scheme 7. It is difficult to say which of these is the more
likely; the one on the left is, in part, effectively the reverse of Scheme 5, whereas the one
on the right is similar to the mechanism that must oocur in the synthesis of II-HE
(Scheme 2). The two routes converge on a common oxythioiminium ion, which could
be in equilibrium with a variety of compounds containing covalently-linked chlorine, as
shown. One or other of these must be responsible for the 272/274 peaks in the mass
spectrum, but none of them is as nicely compatible with the chemical shifts in the
NMR spectrum as the oxythioiminium ion itself. (We can immediately rule out the
structure without a CH2O group, for instance.) As mentioned above, the minor prod-
uct has a doublet at 7.83 ppm, and this is practically identical with that shown by the
dithioiminium ion (7.81 ppm) prepared in unequivocal manner by treatment of II-HE
with perchloric acid, as shown in Scheme 8. (It is rather too high a chemical shift to
belong to a proton ortho to chlorine; for example in N-(2-hydroxyethyl)-2-chloro-5-
nitrobenzylamine, the precursor to II-HE, this is at 7.53 ppm.) Thus it appears that the
minor product exists predominantly in the ionic form in solution, even in a solvent as
nonpolar as chloroform.
Conclusion
The fact that II-HE is rapidly hydrolyzed (by the mechanism of Scheme 5) makes
it more remarkable that the compound shows no reactivity at all toward chymotrypsin.
If it were to bind in the enzyme’s active site, one could envisage the dithiocarbamate
group being held in close proximity to the nucleophilic Ser-195 residue, just as it finds
itself near to the neighboring ethoxy group in Scheme 5. Evidently, however, this sce-
nario fails to lead to reaction, supporting the idea mentioned above that the oxyanion
hole’s inability to interact properly with a sulfur atom is the key factor.
Hydrolysis of 6-NC
As a test of the likelihood that 6-NC would react with an enzymic nucleophile as
suggested in Scheme 3, its reaction with dilute hydroxide ion was first examined. It

84
TREVOR M. KITSON
SCHEME 7. Mechanistic proposals for how the product of alkaline hydrolysis of II-HE can lead,
upon acidification with HCl, to a compound of formula C10H9ClN2O3S.
was found that a solution of 6-NC does indeed rapidly turn yellow at pH 10.2, as expect-
ed if a p-nitrophenoxide ion is being produced. The λmax is 404 nm, but this changes to
a value of 355 nm when the pH is lowered; the observed pKa of the p-nitrophenol
group involved is 7.6. It is assumed that attack of hydroxide ion will give rise to an
enol as this is likely to be more stable than the tautomeric ketoaldehyde because of
conjugation and intramolecular hydrogen-bonding as shown in Scheme 9. This struc-
ture also suggests that similar hydrogen-bonding stabilises the undissociated phenol
group, explaining its rather high pKa value compared to that of p-nitrophenol itself

REPORTER GROUP REAGENTS FOR ENZYMES
85
SCHEME 8. Formation of a dithioiminium ion from II-HE and perchloric acid.
(7.1). (Under more extreme conditions, alkaline hydrolysis of 6-NC involves partial
degradation leading to 2′-hydroxy-5′-nitroacetophenone; 5)
The high pH absorbance at 404 nm is stable, but the low pH peak (at pH 6, for
example) slowly declines in size, by about 50% in an hour at 25°C. This is consistent
with the recyclization of the hydrolysis product as suggested in Scheme 10 (where the
conformation shown—poised for intramolecular 1,4-addition—would be in equilibri-
um with the putative hydrogen-bonded form of Scheme 9), giving back the original
6-NC (which has a λmax of 300 nm but does not absorb at higher wavelengths). At high
pH the phenoxide ion is of course greatly stabilized by delocalization of the nega-
tive charge over the nitro group, but evidently the compound with the undissociated
phenol group is not as stable as the chromone.
The hydrolysis product of 6-NC was added to chymotrypsin at pH 8 and then the
enzyme was passed down a gel filtration column: Fig. 3 shows there is no color asso-
ciated with the protein fraction. Clearly the enzyme and the hydrolysis product do not
have any particularly strong affinity for each other, confirming that in the results dis-
cussed below we are dealing with covalently modified enzyme.
Reaction of 6-NC with Chymotrypsin
Incubation of chymotrypsin with 6-NC at pH 8 over a period of several hours led
to the development of a much more intensely yellow solution than that formed by
spontaneous hydrolysis of 6-NC in the absence of enzyme. After gel filtration of the
reaction mixture, the protein fraction was found to be still yellow at high pH, with λmax
SCHEME 9. Proposed stable conformation of the hydrolysis product of 6-NC (at low pH).

86
TREVOR M. KITSON
SCHEME 10. Proposed cyclisation of certain p-nitrophenol derivatives giving rise to 6-NC. Loss of
HX signifies dehydration or it can represent the elimination of a covalently-linked enzymic group.
of 405 nm, and to absorb at 361 nm at low pH (see Fig. 4); the pKa associated with the
change in maximal wavelength is 7.9 as shown in Fig. 5. The magnitude of the
absorbance of the covalently-linked p-nitrophenoxide ion indicated that about 66% of
the enzyme was modified. (It seems from a variety of experiments that the chy-
motrypsin used in this work is at the most 70–77% pure active enzyme; 18.) These
observations are entirely consistent with the reaction proposed in Scheme 3.
Following modification of chymotrypsin by 6-NC, the enzyme activity (toward
p-nitrophenyl acetate) was measured; three independent experiments gave values of 93,
86, and 97% of the control activity of untreated enzyme. In stopped-flow experiments
with the same substrate, the modified enzyme showed a burst of p-nitrophenoxide ion
production that had 86% of the magnitude of the control and a burst rate constant that
was 93% of the control. It is concluded with certainty therefore that 6-NC does not
react with the catalytically essential nucleophilic group of chymotrypsin (Ser-195) as
this of course would result in complete inactivation. It is also clear that the reporter
group must be in an “ordinary” environment since its pKa and λmax values are insignif-
icantly different from those of the free hydrolysis product of 6-NC; again this suggests
FIG. 3. The UV/visible absorbance of the hydrolysis product of 6-NC at pH 8.0 is shown in
(a). Spectrum (b) is that of the protein fraction after a mixture of chymotrypsin and the hydrolysis product of
6-NC at pH 8.0 had been passed down a gel filtration column.

REPORTER GROUP REAGENTS FOR ENZYMES
87
FIG. 4. Spectra (a) and (b) show the UV/visible absorbance of the reporter group supplied to chy-
motrypsin by 6-NC at pH 10.22 and pH 6.02, respectively. Spectra (c) and (d) are the equivalent results
obtained after the solutions had stood overnight at room temperature.
that the reagent does not react at chymotrypsin’s active site. With I, on the other hand,
the more extensive loss of activity and the grossly perturbed pKa value are consistent
with the idea that the label is in the enzyme’s hydrophobic substrate-binding pocket,
attached to Ser-195(1). As stated before, I and 6-NC are of very similar size and
shape, and both are nucleophilically attacked at position 2. However, with I, the tran-
sition state leading to the tetrahedral intermediate en route to acylation of the
enzyme is exactly analogous to that of a substrate such as p-nitrophenyl acetate, but
with 6-NC the negative charge develops on a different part of the molecule (see
Schemes 1 and 3) and evidently cannot be effectively stabilized by the enzyme’s oxyan-
ion hole. Thus it appears that the same factor explains the lack of reaction of both II-HE
and 6-NC at Ser-195. It is significant that a wide variety of recently developed active-
site-directed inactivators of chymotrypsin all involve the attack of Ser-195 directly on a
carbonyl group (6, 19–21); one exception involves an epoxide instead (22).
As observed with the hydrolysis product of 6-NC, the chymotrypsin-linked reporter
group is stable in the anionic form but not in the neutral form; representative spectra
are shown in Fig. 4. It is concluded therefore that the label detaches from the protein
at low pH by reversal of the modification reaction, as shown in Scheme 10. This was
confirmed by incubating labelled enzyme overnight at room temperature and pH 8.0
(where the reporter group is approximately 50% ionized) and then subjecting it again
to gel filtration at the same pH; taking into account the resulting dilution of the solu-
tion, the UV/visible absorbance showed that 40% of the label had been lost from the
FIG. 5. Ionisation curve for the p-nitrophenol reporter group supplied to chymotrypsin by 6-NC. The
figure shows the best-fit curve calculated by Enzfitter (23); the pKa is 7.9.

88
TREVOR M. KITSON
enzyme under these conditions. If this is released in the form of 6-NC as suggested in
Scheme 10, it would be expected then to undergo slow spontaneous hydrolysis, and
indeed the small molecule fraction of the eluate in the experiment just referred to was
observed to be yellow.
Conclusion
6-NC does react as envisaged with chymotrypsin to provide a conveniently moni-
torable reporter group, but it involves a nonessential enzymic nucleophile and proba-
bly occurs on the outer surface of the enzyme molecule. The label is chemically sta-
ble only above its pKa (7.9) and therefore identifying the amino acid residue that is
modified may prove to be problematic.
ACKNOWLEDGMENTS
I am grateful to Mr G. H. Freeman for the nitration of chromone and to Dr P. J. B. Edwards for recording
NMR spectra.
REFERENCES
1. Kitson, T. M., and Freeman, G. H. (1993) Bioorg. Chem. 21, 354–365.
2. Kitson, T. M., and Kitson, K. E. (1994) Biochem. J. 300, 25–30.
3. Kitson, T. M., and Kitson, K. E. (1995) in Enzymology and Molecular Biology of Carbonyl Metabolism
5 (Weiner, H., Holmes, R. S., and Wermuth, B., Eds.), pp. 35–43, Plenum Press, New York/London.
4. Joule, J. A., Mills, K., and Smith, G. F. (1995) Heterocyclic Chemistry, 3rd ed., Chapman & Hall, London.
5. Ellis, G. P., and Thomas, I. L. (1973) J. Chem. Soc. Perkin 1, 2781–2785.
6. Pochet, L., Doucet, C., Schynts, M., Thierry, N., Boggetto, N., Pirotte, B., Jiang, K. Y., Masereel, B., de
Tullio, P., Delarge, J., and Reboud-Ravaux, M. (1996) J. Med. Chem. 39, 2579–2585.
7. Thorn, G. D., and Ludwig, R. A. (1962) The Dithiocarbamates and Related Compounds, Elsevier, Amsterdam/
New York.
8. Hart, B. W., and Faiman, M. D. (1992) Biochem. Pharmacol. 43, 403–406.
9. Kitson, T. M. (1977) J. Chem. Soc. Perkin 1, 565–566.
10. Hirohara, H., Bender, M. L., and Stark, R. S. (1974) Proc. Nat. Acad. Sci. USA 71, 1643–1647.
11. Asboth, B., and Polgar, L. (1983) Biochemistry 22, 117–122.
12. Asboth, B., Stokum, E., Khan, I. U., and Polgar, L. (1985) Biochemistry 24, 606–609.
13. Kitson, T. M. (1978) Biochem. J. 175, 83–90.
14. Kitson, T. M. (1980) J. Chem. Soc. Perkin 1, 1724–1725.
15. Beak, P., and Becker, P. D. (1982) J. Org. Chem. 47, 3855–3861.
16. Scully, F. E., and Ortega, T. (1989) J. Org. Chem. 54, 2978–2980.
17. Lown, J. W., and Chauhan, S. M. S. (1983) J. Org. Chem. 48, 3901–3908.
18. Kitson, T. M. (1998) Biochim. Biophys. Acta 1385, 43–52.
19. Gutscow, M., and Neumann, U. (1995) Monatshefte fur Chemie 126, 1145–1149.
20. Kim, D. H., and Ryoo, J. J. (1995) Bioorg. Med. Chem. Lett. 5, 1287–1292.
21. Li, M., Luo, W., and White, E. H. (1995) Arch. Biochem. Biophys. 320, 135–140.
22. Kim, Y. J., Li, Z.-H., Kim, D. H., and Hahn, J. H. (1996) Bioorg. Med. Chem. Lett. 6, 1449–1452.
23. Leatherbarrow, R. J. (1987) Enzfitter Manual, Elsevier, Amsterdam.